Moisture conditions
in timber frame roof and wall structures
Stig Geving and Sivert Uvsløkk
273 Project report 2000
Test house measurements for verifi cation of
heat-, air and moisture transfer models
BYGGFORSK
Norwegian Building Research Institute
Project report 273 − 2000
Test house measurements for verifi cation of heat-, air and moisture transfer models
Moisture conditions
in timber frame roof and wall structures
Stig Geving and Sivert Uvsløkk
Project report 273
Stig Geving and Sivert Uvsløkk
Moisture conditions in timber frame roof and wall structures
Test house measurements for verifi cation of heat-, air and moisture transfer models
Key Words: moisture, test house, timber frame, moisture transport, walls, roofs, weather station, water vapour permeability, sorption curves, air tightness
ISSN 0801-6461 ISBN 82-536-0700-8 100 eks. printed by S.E. Thoresen as Content:100 g Kymultra Cover: 200 g Cyclus
© Norwegian Building Research Institute 2000
Address: Forskningsveien 3 B Postboks 123 Blindern N-0314 Oslo
Phone: +47 22 96 55 55
Fax: +47 22 69 94 38 and +47 22 96 55 42
PREFACE
This report presents a part of the work that was carried out within the Strategic Research Programme «Moisture in Building Materials and Constructions». The programme was carried out in a four year period 1993-1997 as a cooperation between the Norwegian Building
Research Institute (NBI) and Department of Building and Construction Engineering, Norwegian University of Science and Technology (NTNU). The programme was mainly funded by the Norwegian Research Council, and additionally by internal funding from the participating institutions.
The programme included the following projects:
0. General activities and programme management 1. Literature survey
2. Moisture physics 3. Calculation programs 4. Material properties 5. Verification
6. Dr.ing. (PhD) studies 7. International cooperation
The work presented in this report has been carried out within project 5 "Verification". The test house measurements were made in the period 1994-98. Since 1998 extensive analysis work and additional measurements have been performed.
Trondheim, March 2000
Stig Geving Sivert Uvsløkk
CONTENTS
1 INTRODUCTION 5
2 DESCRIPTION OF THE TEST HOUSE 6
2.1 General 6
2.2 Logging system 8
2.3 Heating and ventilation 8
2.4 Automatic weather station 9
3 DESCRIPTION OF WALL AND ROOF SECTIONS 14
3.1 Wall sections 14
3.2 Roof sections 18
3.3 Materials used in wall and roof sections 21
3.4 Hygroscopic material properties 21
3.5 Air tightness measurements 23
4 HYGROTHERMAL MEASUREMENTS 25
4.1 General 25
4.2 Timber frame walls 28
4.3 Timber frame roofs 29
4.4 Sections of aerated concrete 29
4.5 Investigation of forced convection in timber frame walls 29
4.6 Inspection of wall elements 29
4.7 Control of the method for measuring moisture content 30
4.8 Indoor climate 33
4.9 Publications where the measurements have been presented 34
5 RESULTS 35
5.1 General 35
5.2 Diagrams – eastern wall sections 36
5.3 Diagrams – western wall sections 39
5.4 Diagrams – roof sections 43
6 DISCUSSION 47
6.1 Wall sections 47
6.2 Roof sections 48
7 CONCLUSIONS 49
8 REFERENCES 50
1 INTRODUCTION
As a part of the research programme "Moisture in Building Materials and Constructions"
(1993-97), experiments have been performed on different building envelope structures at a test house in Trondheim, Norway. Most of the test elements were lightweight timber frame constructions. The roof elements were all horizontal with a high degree of one-
dimensionality. The timber frame walls had different combinations of vapour retarders and wind barriers. The external surfaces were exposed to the ambient climate, while the indoor climate in the house was controlled at 23 °C and 50 % RH. The outdoor climate was
monitored by an automatic weather station situated 17 meters from the test house. Moisture and temperature conditions in the elements were monitored continuously in the period 1994- 98. Since then extensive analysis work and additional measurements have been performed.
The main purposes of the test house measurements were:
1. Collecting data for comparison with computer simulations of transient moisture transfer in building constructions. The measurement data from this test house are made available for researchers who want to verify computer models for heat, air and moisture transfer through building structures. The data are available on digital format (CD-ROM).
2. Investigate how untraditional combinations of indoor and outdoor vapour resistance (vapour barrier and wind barrier) influence the moisture conditions of a timber frame wall.
This report gives a description of the data that are available for verification purposes from the test house. The report comprises descriptions of :
• Test house
• Wall and roof elements investigated
• Instrumentation of the elements
• Boundary conditions
• Outdoor climatic parameters measured
• Measured material properties
• Some measurement results from wall and roof elements
2 DESCRIPTION OF THE TEST HOUSE 2.1 General
The test house is located on a field station belonging to the Norwegian Building Research Institute and Department of Building and Construction Engineering, NTNU. The field station is located on an open field at Voll (Jonsvannsveien 159) in Trondheim, approximately 6 km south-east of the centre of the city. The exact location is N63°25' E10°28'. The field station consists of a test house with removable roof and wall elements (which is described in this report), another test house which can be rotated for wind pressure studies, an automatic weather station (also described in this report) and a small measurement house in connection with the weather station, see figure 1. The test house is shown in Figure 2.
The roof and facades of the test house consist of prefabricated sections fixed to the outside of a steel frame structure, see figure 3. The test house is orientated in north-south direction and has the following indoor dimensions: length 10.7 m, width 3.45 m and height 3.5 m. The roof sections span from facade to facade, and have a 1:40 slope. All sections are 1,2 m wide and they are separated from each other regarding air and moisture transfer, by use of polyethylene foil. The sections may be changed individually without disturbing the neighbour sections.
There are a total of 16 wall sections on the western and eastern facades and 8 roof sections, see figure 4. The test house is equipped with a low temperature electric floor heating system, balanced mechanical ventilation with heat recovery and an automatic air humidifying system.
Figure 1 A map showing the field station
Test house
Automatic weatherstation
Rotating Jonsvannsveien
N
W E
S
Figure 2 The northeast side of the test house.
Figure 3 East-west section of the test house showing how the elements are fixed to the steel frame structure.
Walltestelements Steelframe
Rooftestelements
4750mm 3650mm
230 mm 3245 mm
230 mm100 mm
EAST WEST
Figure 4 Plan of the test house in Trondheim, showing the location of the wall and roof test elements.
2.2 Logging system
The logging system is a combined system of loggers and communication software on PC.
Three loggers (Campbell CR 10) are each connected with the measurement sensors trough a multiplexer (Campbell AM 32 B), see figure 5. For the temperature measurements a common reference was used for each multiplexer. The loggers each have an internal temperature reference, but they were also connected with an external temperature reference (Campbell 10TCRT Thermocouple Reference).
Because of very high electrical resistance (low levels of electrical current) coaxial cables were used for the moisture measurements to avoid that electrical noise from other electrical cables and equipment influenced the measurement signals. To convert the measurement signals (voltage) to moisture content a converter (Delmhorst MT(G) 40) was inserted between the loggers and the multiplexers. This conversion is further described in chapter 4.1. The logging system is more thoroughly described and documented in (Homb, 1998).
2.3 Heating and ventilation
The test house is equipped with a low temperature electric floor heating system, i.e. a Flexel Mark III heating foil with maximum power output of 90 W/m2. The heating foil is automatically regulated through a thermostat with a temperature sensor in the room air.
Throughout the measurement period the temperature of the room was maintained at approximately 23 °C.
The house is ventilated with a balanced mechanical ventilation system (Covent Master CM1) with heat recovery and an automatic air humidifying system. The air flow rate can be controlled between 60-200 m3/h. The ventilation system will normally give a constant air change rate of 0,5 m3/m3h. The supply air is heated to the setpoint temperature of the room with an air heater battery. The air supply to the room is through a spiral wound tube with a diameter of 160 mm in the whole length of the house. Holes have been made in the tube to
The air humidifying system (Steamatic) automatically adds water vapour to the supply air to maintain the wanted level of relative humidity in the room air. Throughout the measurement period the RH of the room was maintained at approximately 50 %.
Figure 5 Connection of measurement sensors to one of the three loggers.
2.4 Automatic weather station
Outdoor climatic data are measured by a Milos 500 Vaisala automatic weather station (AWS) located 17 m to the south of the test house, see Figure 1. A picture of the weather station is shown in Figure 6. Hourly values of air temperature, relative humidity, air pressure, wind speed, wind direction, precipitation, global radiation and longwave radiation are recorded.
Daily average values of outdoor air temperature, relative humidity and global radiation for the measurement period is shown in figure 7. Occasionally, when weather data are missing they are reconstructed from three other local weather stations (respectively 1, 2 and 40 km from the test house). The sensors used in the Milos 500 Vaisala AWS are shown in Table 1.
Campbell logger CR 10
Delmhorst MT (G) 40 Converter
Campbell MUX AM 32 B
Delmhorst MT (G) 40 Converter
Campbell MUX AM 32 B
Moisture sensors, max 32 channels
Moisture sensors, max 32 channels
Campbell MUX
AM 32 B Temperature sensors,
max 64 channels
Campbell
Temperature reference 10TCRT
Figure 6 The automatic weather station .
Table 1 Sensors used in the Milos 500 Vaisala AWS
Type Trade name
Airtemperatureandhumiditysensor VaisalaHMP35D
Pressuresensor VaisalaDPA21
Wind speed sensor Vaisala WAA 15A
Wind direction sensor Vaisala WAV 15A
Precipitationamount Geonor
Precipitationdetector VaisalaDRD11A Solar radiation pyranometer Kipp & Zonen CM 6B
Pyrgeometer Kipp & Zonen CG 1
The AWS has been operating since 15. March 1995. From 1. March 1996 the Norwegian Meteorological Institute took over the operation of the AWS, ensuring a good quality of the measurements. Since then there is only missing data for a few and short periods. Before 1.
March 1996 there were several periods when the AWS was not working. The most serious breakdown took place in the period 15. October 1995 - 29. February 1996 because of serious vandalism on the weather station.
Table 2 are shown the format of the meteorological parameters measured and recorded
Table 2 Format of the meteorological parameters measured and recorded since 1. March 1996
No. Symbol Parameter Unit
1 TT Temperature, last minute last hour °C
2 TTM Temperature,averagelasthour °C
3 TTN Temperature, minimum last hour °C
4 TTX Temperature,maximumlasthour °C
5 UU Relative humidity, last minute last hour %
6 UUM Relativehumidity,averagelasthour %
7 FF Wind speed, last 10 minute average last hour m/s
8 FM Wind speed, average last hour m/s
9 FG Windspeed,3secondsmax.gustlasthour m/s
10 FX Wind speed, max. 10 minute running average last hour m/s 11 DD Winddirection,belongstoFF(0° =north,90° =east,etc) °
12 DM Winddirection,belongstoFM °
13 DX Winddirection,belongstoFX °
14 RA Totalcontentinraingauge mm
15 RR Increase in rain gauge last hour mm
16 RT Numberofminutesprecipitationlasthour 0-60
17 PO Air pressure in height of station, last minute last hour hPa 18 POM Air pressure in height of station, average last hour hPa 19 PON Airpressureinheightofstation,minimumlasthour hPa 20 POX Air pressure in height of station, maximum last hour hPa 21 PP Airpressure,reducedtosealevel,standardformulae hPa 22 PF Air pressure, reduced to sea level regarding temperature hPa
23 PT Air pressure difference, 3 hours hPa
24 AA Airpressurecharacteristics,3hours(SYNOP-code) 0-8
25 QO Global radiation, accumulated last hour W/m2
26 QOX Globalradiation,maximumlasthour W/m2
27 QL Longwaveradiation,accumulatedlasthour W/m2
28 QLX Longwave radiation, maximum last hour W/m2
29 SS Snowdepth,lastminutelasthour cm
The AWS measures most parameters every five seconds (wind every second). Averages over 60 seconds are calculated (minute values), and these are used to find the hourly values described in Table 2. The hourly values are generated at every shift of hour and consist of:
• Instantaneous value = minute average last minute before shift of hour
• Maximum value = maximum minute average last hour
• Minimum value = minimum minute average last hour
• Average value = average of all minute values last hour
Regarding longwave radiation it should be noted that it is the longwave downward radiation to the instrument (Ldown) that is measured (QL and QLX in Table 2), and not the net longwave radiation (Lnet). Ldown is calculated with the following formulae:
L V
K T
down = +5 67 10. ⋅ −8⋅ s4
where V is output of the pyrgeometer (V), K is a calibration factor (Vm2/W) and Ts is the sensor temperature (K). If net longwave radiation (Lnet) is wanted the longwave radiation part
of the formula above should be subtracted for the specific medium. For example for the sensor Lnet is calculated as follows:
Lnet = Ldown −5 67 10. ⋅ −8⋅Ts4
Climatic data for the period before 1. March 1996 are available, but fewer parameters are available for that period. The period that is available is 1. October 1994 - 1. March 1996. It should be noted that during this period the AWS partly had not started measuring and partly was not working. That means that for some periods we had to reconstruct data from other climatic stations, as shown in Table 3.
Table 3 Main climatic stations used for different periods
Period Climatic station Comments 01.10.94-15.03.95 MoholtAWS
15.03.95-01.09.95 VollAWS Manysmallholesinmeasurements 01.09.95-01.03.96 Moholt AWS
01.03.96- Voll AWS Operated by DNMI, few holes in measurements
For the period 01.10.94-30.09.95 climatic data from several other stations or measurement sites were employed when data was missing from both Voll AWS and Moholt AWS. These stations/measurement sites were as follows: 1) Risvollan AWS, 2) Værnes airport (main observations by DNMI), 3) Stokkanhøgda (global radiation at a single-family house) and 4) Surface temperatures from test house Voll.
In addition to climatic data for the period 01.03.96 and until today, which are available on the format described in Table 2, climatic data for the whole period 01.10.94 and until today are available on a simplified format described in Table 4. This format that contain only a few parameters have been used to generate climatic files for different heat-, air- and moisture programs.
Table 4 Format of simplified hourly climatic data available for the whole period 01.10.94 and until today (available on EXCEL-format).
No. Parameter Unit
1 Temperature,averagelasthour °C
2 Relative humidity, average last hour % 3 Global radiation, accumulated last hour W/m2
4 Windspeed,averagelasthour m/s
5 Cloudcover(1-8)* -
*ThevalueiseitherpickedfromVærnesairport(01.10.94-30.09.95)orchosenequalto5.
In addition to global radiation, most heat-, air- and moisture programs also require values for diffuse- and direct solar radiation. A program ("vollfil") was developed that estimated the hourly values of diffuse and direct solar radiation from global radiation. This was based on a method described in (Skartveit and Olseth, 1988). The program reads a climatic file on the format described in Table 4 and is then able to generate climatic files including diffuse- and direct solar radiation on different formats.
Figure 7 Outdoor climate measured at the automatic weather station (daily averages).
-20 -10 0 10 20
1.jan.95 2.jul.95 1.jan.96 1.jul.96 31.des.96 1.jul.97 31.des.97 1.jul.98 Temperature(o C)
0 20 40 60 80 100
1.jan.95 2.jul.95 1.jan.96 1.jul.96 31.des.96 1.jul.97 31.des.97 1.jul.98
RelativeHumidity(%)
0 100 200 300 400
1.jan.95 2.jul.95 1.jan.96 1.jul.96 31.des.96 1.jul.97 31.des.97 1.jul.98 Date
GlobalRadiation(W/m2 )
3 DESCRIPTION OF WALL AND ROOF SECTIONS 3.1 Wall sections
The overall dimensions of the wall sections are 1190 mm x 3250 mm, with stud and
top/bottom plate dimensions of 48 mm x 148 mm (spruce), except of section E8 with 300 mm wood I-joists (46 mm x 46 mm wood flanges, 10 mm particle board web). To achieve normal conditions for natural convection, the height of the insulated cavities was made normal (2418 mm) by mounting an extra “sill plate” in the wall sections approximately 0.9 m above the floor level as shown in figure 8. The “E” and “W” sections were located on the eastern and western facades respectively. A PE-foil isolated the test wall perimeter from the surrounding constructions in terms of mass transfer, and adiabatic conditions were maintained at the wall perimeter with respect to heat transfer.
The wall sections are described in detail in table 5 and figures 9 -10. All the timber frame walls have 150 mm glass fibre insulation (density ≈ 18 kg/m3, thermal conductivity = 0,036 W/mK), except section E8 which has 300 mm insulation and a timber frame made of wooden I-joists instead of solid spruce. All the wall sections, except sections W2, W5, E5 and E6, have a 23 mm ventilated air gap and 19 mm shiplap cladding outside of the wind barrier.
Section W2 has 19 mm shiplap cladding but no air gap between the wind barrier and the cladding. Section W6 has two noggings with 0.8 m spacing).
Table 5 Description of materials used in the various wall sections. W1 and E1 denote section no. 1 on the western facade and eastern facade respectively.
Secti ons
Internallining Vapourbarrier Wallconstruction/insulation Windbarrier W1 woodfibreboard
12mm
polypropylenefoil (windbarrier)
timberframewall(48x148mm)/ 150mmglassfibre
woodfibreboard 3mm W2*+
W3
woodfibreboard 12mm
polypropylenefoil (windbarrier)
timberframewall(48x148mm)/ 150mmglassfibre
gypsumboard 9mm W4+
E4
woodfibreboard 12mm
- timberframewall(48x148mm)/ 150mmglassfibre
gypsumboard 9mm W5+
E5
- - LECA-block250mm/ 80mm
PUR-insulationinthemiddle
- W6 gypsumboard
9mm
- timberframewall(48x148mm) withnoggings/150mmglassfibre
gypsumboard 9mm W7+
W8
gypsumboard 9mm
- timberframewall(48x148mm)/ 150mmglassfibre
gypsumboard 9mm E1 woodfibreboard
12mm
polyethylenefoil 0.15mm
timberframewall(48x148mm)/ 150mmglassfibre
asphaltimpr.wood fibreboard,12mm E2** woodfibreboard
12mm
polyethylenefoil 0.15mm
timberframewall(48x148mm)/ 150mmglassfibre
spunbonded polyethylenefoil E3** woodfibreboard
12mm
polyethylenefoil 0.15mm
timberframewall(48x148mm)/ 150mmglassfibre
polypropylenefoil E6+
E7
- - aeratedconcrete
300mm
- E8 woodfibreboard
12mm
- timberframewall(300mmI-joists) /300mmglassfibre
gypsumboard 9mm
*Noairgapbetweenthewindbarrierandthecladding.
**InSeptember1996thewallsectionsE2andE3weremodified.ThewindbarrierofsectionE3waschangedto 9mmgypsumboard.Forbothsectionsahorizontal3mmwidegapwassawnthroughtheinternalliningand vapourbarriernearthebottomsill,andasimilargapthroughthewindbarriernearthetopsillattheexternalside.
Figure 8 Dimensions of the timber frame wall sections, in mm. The location of the sensors for measuring moisture content (· ·), and temperature (°) to the right.
One of the purposes of the measurements was to investigate how untraditional combinations of indoor and outdoor water vapour resistance influence on the moisture conditions of timber frame walls. In table 6 is therefore shown the water vapour resistance on cold and hot side and the ratio, for the timber frame walls. The water vapour resistances on the cold side were probably somewhat lower during the winter than the values given in table 6. This because the water vapour resistance of a material is dependent on the moisture content, especially for wood based materials. During the winter the RH for the wind barriers were probably higher than 72 %, which is the average RH-level for the cup-measurements of which table 6 is based on.
wood fibre board, 12 mm
polypropylene foil (wind barrier) glass fibre insulation, 150 mm wood fibre board, 3 mm wood fibre board, 12 mm
polypropylene foil (wind barrier) glass fibre insulation, 150 mm gypsum board, 9 mm
wood fibre board, 12 mm
polypropylene foil (wind barrier) glass fibre insulation, 150 mm gypsum board, 9 mm
wood fibre board, 12 mm glass fibre insulation, 150 mm gypsum board, 9 mm
rendering
LECA-block, 250 mm (total, including PUR) PUR insulation, 80 mm
rendering
gypsum board, 9 mm
glass fibre insulation with noggins, 150 mm gypsum board, 9 mm
gypsum board, 9 mm
glass fibre insulation, 150 mm gypsum board, 9 mm
gypsum board, 9 mm
glass fibre insulation, 150 mm gypsum board, 9 mm
Figure 9 Wall sections of the western facade.
wood fibre board, 12 mm polyethylene foil, 0,15 mm glass fibre insulation, 150 mm
asphalt impregnated wood fibre board, 12 mm wood fibre board, 12 mm
polyethylene foil, 0,15 mm glass fibre insulation, 150 mm
spunbonded polyethylene foil (wind barrier) wood fibre board, 12 mm
polyethylene foil, 0,15 mm glass fibre insulation, 150 mm polypropylene foil (wind barrier) wood fibre board, 12 mm
glass fibre insulation, 150 mm gypsum board, 9 mm
rendering
LECA-block, 250 mm (total, including PUR) PUR insulation, 80 mm
rendering
aerated concrete, 300 mm
aerated concrete, 300 mm
wood fibre board, 12 mm
glass fibre insulation, 150+150 mm gypsum board, 9 mm
Figure 10 Wall sections of the eastern facade.
Table 6 Water vapour resistance (diffusion equivalent air layer thickness) on hot and cold side of the timber frame walls. "Ratio" is water vapour resistance on the hot side divided by the water vapour resistance on the cold side.
Wall section
Watervapourresistance, diffusionequivalentair
layerthickness,sd
[m] [m] [-]
No Hotside Coldside Ratio
W1 5,0 0,14 35
W2 5,0 0,08 62
W3 5,0 0,08 62
W4 0,53 0,08 7
W6 0,08 0,08 1
W7 0,08 0,08 1
W8 0,08 0,08 1
E1 64 0,23 280
E2 64 0,02 3200
E3 64 4,4 14
E4 0,53 0,08 7
E8 0,53 0,08 7
3.2 Roof sections
The overall dimensions of the roof sections were 1190 mm x 4750 mm, as shown in figure 11.
The roof sections are described in detail in table 7 and figure 12. A PE-foil isolated the roof section perimeter from the surrounding sections in terms of mass transfer, and adiabatic conditions were maintained at the test section perimeter with respect to heat transfer.
The timber frame roof sections (R1-R6) consists of the following materials from the interior;
12 mm chipboard, 0,15 mm polyethylene foil, 200 mm glass fibre insulation (48x198 mm wooden rafters), 22 mm plywood and 1,3 mm PVC roofing membrane. In two of the timber frame roof sections (R5 and R6) the rafter spacing is 1.2 m to achieve 1-dimensional conditions in the middle of the section, while the other four timber frame roofs (R1-R4) have a more normal 0.6 m rafter spacing. In sections R5 and R6 the wooden rafters are separated from the rest of the section with a polyethylene foil (see Figure 12 and 18). The sections R7 and R8 consist of 300 mm and 200 mm aerated concrete respectively. R8 has a 100 mm layer of rockwool fibre insulation on the top.
Table 7 Description of materials used in the various roof sections.
Sections Internal lining
Vapourbarrier Roofconstruction/
insulation
Roofing R1 chipboard
12mm
polyethylene foil,0.15mm
timberframe(48x198mm)/ 200mmglassfibre
22mmplywood/ 1.3mmPVCroofing membrane(darksideupwards) R2+R3
+R4
chipboard 12mm
polyethylene foil,0.15mm
timberframe(48x198mm)/ 200mmglassfibre
22mmplywood/ 1.3mmPVCroofing membrane(lightsideupwards) R5 chipboard
12mm
polyethylene foil,0.15mm
timberframe*/ 200mmglassfibre
22mmplywood/ 1.3mmPVCroofing membrane(lightsideupwards) R6 chipboard
12mm
polyethylene foil,0.15mm
timberframe*/ 200mmglassfibre
22mmplywood/ 1.3mmPVCroofing membrane(darksideupwards)
R7 - - aeratedconcrete
300mm
1.3mmPVCroofingmembrane(dark sideupwards)
R8 - - aeratedconcrete 1.3mmPVCroofingmembrane(dark
Four elements (R1, R6-R8) have a dark coloured roofing membrane (emissivity ≈ 0.9), while the other four have the same roofing membrane, but the light side facing upwards (emissivity
≈ 0.65).
Figure 11 Dimensions of the timber frame roof sections R1-R6, in mm. The location of the sensors for measuring moisture content (· ·), and temperature (°) at the top.
PVC roofing membrane (dark), 1,3 mm plywood, 22 mm
polyethylene foil, 0,15 mm chipboard, 12 mm
PVC roofing membrane (light), 1,3 mm plywood, 22 mm
polyethylene foil, 0,15 mm chipboard, 12 mm
PVC roofing membrane (light), 1,3 mm plywood, 22 mm
polyethylene foil, 0,15 mm chipboard, 12 mm
PVC roofing membrane (light), 1,3 mm plywood, 22 mm
polyethylene foil, 0,15 mm chipboard, 12 mm
PVC roofing membrane (light), 1,3 mm plywood, 22 mm
polyethylene foil, 0,15 mm chipboard, 12 mm
PVC roofing membrane (dark), 1,3 mm plywood, 22 mm
polyethylene foil, 0,15 mm chipboard, 12 mm
PVC roofing membrane (dark), 1,3 mm aerated concrete, 300 mm
PVC roofing membrane (dark), 1,3 mm rockwool insulation, 50 mm + 50 mm aerated concrete, 200 mm
12 Roof sections. The timber frame sections are insulated with 200 mm glass fiber
3.3 Materials used in wall and roof sections
The trade name, manufacturer in Norway, thickness and density of the materials used in the test sections are given in Table 8. All the wall and roof sections (except E5 and W5) were all prefabricated indoors in a partly heated environment. This means that all the materials that were used were relatively dry at installation. It is assessed that the various materials were at moisture equilibrium at approximately 60% RH, which can be used as initial moisture conditions. At installation of the elements the moisture content of the wooden parts (studs, top/bottom sills) were measured to be approximately 11 weight%.
Table 8 Specification of materials used in the test house
Material TradenameinNorway/ Thickness Density manufacturer * [mm] [kg/m3] Aeratedconcrete "Siporexvegg/takelement"/
Siporex A/S, Norway - 474
Asphalt impregnated wood fibreboard,porous
"Asfalt vindtett" / HuntonFiberA/S,Norway
12 251
Glass fibre insulation "Glava matte A 36"/
Glava A/S, Norway 100/150 18
Gypsumboard,exteriorgrade "GU,Utvendiggipsplate"/ NorgipsA/S,Norway
9 757
LECA-block "LECA-isoblokk" /
as Norsk Leca 250
Plywood ** "WBPP-30,Vänerplytakplater"/
Vänerply AB, Sweden 22 411
Polyethylene foil, 0.15mm (PE-buildingmembrane)
"Tenotett" / RosenlewAB,Sweden
0.15 Spunbonded polypropylene
foil, wind barrier "Rockwool airbarrier" /
Reemay Inc., USA 0.27
PVC roofing membrane "Sarnafil SE 3" /
Protan A/S, Norway 1.3 Rockwoolinsulation "Rockwool"/
a/sRockwool,Norway
50+50 ∼ 150+
∼ 90
Norway spruce (Picea abies) 350-465
Spunbonded polyethylene foil,
wind barrier "Isola vindsperre (Tyvek)" /
Du Pont de Nemours, Luxemb. 0.14 Woodchipboard "Sponplate"/
Norske skogindustrier A/S 12 554 Wood fibre board
(hardboard)
"Huntonit bygningsplate" / HuntonFiberA/S,Norway
3 / 11 803
*Thereaderiscautionedthatmanufacturersmaychangeproductsovertimewhileretainingthesamebrandname.
**Materialofspruceorpine,numberoflayers=7.
3.4 Hygroscopic material properties
To be able to simulate the hygrothermal conditions of the various constructions the material properties have to be known. The most important material properties in this context are probably the water vapour permeability and the hygroscopic sorption curves. These two properties have been measured for most of the materials used in the constructions in the test house, and the results are presented in table 9-11. A more detailed presentation of these measurements and results are given in (Bergheim et.al, 1998) and (Geving, Time and Hovde,
Table 9 Water vapour permeability for spruce measured in the transverse direction, i.e. a combined radial/tangential direction, for various RH-levels (with standard deviation σ). The values represent a mean value for thickness ranging from 2.4 mm to 10 mm.
Material Density kg·m-3
Watervapourpermeability±±±± σ (10-12·kg·m-1·Pa-1·s-1)
RH:31% 63% 72%
Spruce * 350-380 1.75 ± 0.34 6.01 ± 0.51
Spruce * 440-465 3.94 ± 0.09 7.23 ± 0.26
*Themeasurementsofthevapourpermeabilityofsprucearemorethoroughlyreportedin[Time,1998].
Table 10 Measured water vapour permeance (with standard deviation σ) and corresponding equivalent air layer thickness. Average RH-level during measurements was 72 % (50%/94%).
Material Thickn.
mm
Density (kg/m3)
Watervapour permeance ±±±± σ (10-12·kg·m-2·Pa-1·s-1)
Equivalentairlayer thickness ±±±± σ
(m)
Aeratedconcrete 21 474 1180 ± 70 0.165 ± 0.01
Asphalt impregnated wood fibre board,porous
12 251 845 ± 264 0.231 ± 0.10
Gypsumboard,exteriorgrade 9 757 2430 ± 80 0.080 ± 0.003
Plywood 22 411 145 ± 10 1.34 ± 0.10
Polyethylenefoil,vapourbarrier 0.15 3.09 ± 0.57 63.0 ± 14 Polypropylenefoil,windbarrier 0.27 44.1 ± 3.3 4.4 ± 0.4
PVCroofingmembrane 1.3 15.0 ± 1.1 13.0 ± 1.0
Spunbondedpolyethylenefoil, windbarrier
0.14 9720 ± 760 0.020 ± 0.002
Woodfibreboard,highdensity 11 803 371 ± 25 0.525 ± 0.04
Table 11 Measured equilibrium moisture contents (weight-%) for adsorption and desorption curves.
Wood chipboard
Woodfibreboard (hardboard)
Plywood Aeratedconcrete Spruce
RH(%) Ads. Des. Ads. Des. Ads. Des. Ads. Des. Ads. Des.
11.3 - - - 3.4 - 3.8 - 1.01 - 3.1
32.9 5.7 - 5.1 7.2 6.8 8.3 0.23 1.40 7.2 7.8
53.5 7.7 - 7.0 8.8 8.9 10.3 0.44 1.59 9.6 10.0
75.4 10.6 - 9.6 12.1 12.2 14.8 0.93 1.89 13.2 14.7
81.2 11.5 - 10.5 - 13.2 - 1.21 - 14.1 -
94 15.6 - 13.7 - 17.8 - 1.88 - 18.6 -
97.4 - - 16.0 16.0 18.9 18.9 2.43 2.43 21.0 21.0
3.5 Air tightness measurements
The air tightness of the various wall sections was measured on site by blowing air through a tube, via an air-flow meter (type Rotameter), into the two insulation cavities of each section.
By simultaneously controlling the indoor pressure of the test house, by use of an adjustable fan installed in the test house door, three types of air leakages were measured at several pressure differences in the range 5 to 50 Pa. The pressure differences between the wall cavity and indoor air (∆Psi) and between indoor and outdoor air (∆Pie) were measured with micro- manometers. The air flow through the interior surface (Qsi), through the exterior surface (Qse) and the sum of these two (Qse+si) were measured. The sum Qse+siwas found by adjusting to zero the pressure difference between indoor and outdoor air, and vary ∆Psi between 5 and 50 Pa. Qse was found by adjusting to zero the pressure difference between the wall cavity and indoor air, and vary ∆Pie between 5 and 50 Pa. Qsi was found by adjusting to zero the pressure difference between the wall cavity and outdoor air, and vary ∆Psi between 5 and 50 Pa. The measurement setup is shown in figure 13.
On basis of these results, air leakage data for the various sections were estimated. It was assumed that there was no air leakage through the perimeter of the test sections or from the upper wall cavities to the "dummy" wall cavities in the bottom of the wall. The air leakage area used in the calculations of specific air leakage is therefore 2,88 m2 (2,418m x 1,19 m) for each face of the wall. The specific air leakage through the wall section (qie), through the interior surface (qsi), through the exterior surface (qse) and the sum of these two (qse+si) were calculated. In figure 14 an example of these results are shown for section E1.
The dependance of specific air leakage to the pressure difference was found by curve fitting, using the following equation:
α Pβ
q= ⋅∆ (1)
where α and β are coefficients dependent of type of leakage and section. In table 12 these coefficients are given for each air leakage type, together with the specific air leakage at 50 Pa.
13 Sketch of air leakage measurements in the test house
Figure 14 Example of results for specific air leakage for section E1.
Table 12 Specific air leakage of the various sections
Wall section
Coefficients α and β according to eq. (1) Specific air leakage at 50 Pa [10-4 ⋅m3/m2s]
qsi+se qse qsi qie qsi+se qse qsi qie
W1 β: 0,884 0,914 1,0 0,963 3,6 4,1 3,4 1,9
α [10-5]: 1,14 1,15 0,686 0,433
W2 β: 0,922 0,98 0,971 0,974 3,5 4,6 2,5 1,6
α [10-5]: 0,943 0,991 0,564 0,359
W4 β: 0,931 0,945 0,946 0,945 3,2 1,0 5,6 0,86
α [10-5]: 0,847 0,252 1,39 0,213
W6 β: 0,935 0,937 0,939 0,937 5,2 0,78 9,9 0,72
α [10-5]: 1,35 0,2 2,52 0,185
W7 β: 1.01 0,863 1,03 0,886 3,3 0,73 5,7 0,65
α [10-5]: 0,636 0,251 1,01 0,204
W8 β: 0,986 0,938 1,04 0,954 3,0 0,94 5,8 0,81
α [10-5]: 0,632 0,241 0,99 0,195
E1 β: 0,94 0,957 0,818 0,879 3,8 4,5 3,0 1,8
α [10-5]: 0,965 1,06 1,22 0,577
E2* β: 0,875 0,802 0,723 0,763 17 17 16 8,3
α [10-5]: 5,71 7,37 9,53 4,18
E3** β: 0,766 0,735 0,732 0,734 13,9 10,7 19,3 6,9
α [10-5]: 6,96 6,02 11 3,89
E4 β: 0,889 0,628 0,882 0,644 4,1 0,41 8,0 0,39
α [10-5]: 1,26 0,351 2,53 0,314
E8 β: 0,891 0,614 0,87 0,651 3,25 0,75 5,8 0,67
α [10-5]: 0,997 0,678 1,94 0,521
Wall section E1
y=5.77E-06x8.79E-01 y=1.06E-05x9.57E-01
y=9.65E-06x9.40E-01
y=1.22E-05x8.18E-01
0.E+00 1.E-04 2.E-04 3.E-04 4.E-04 5.E-04
0 10 20 30 40 50
Pressuredifference,Pa Specificairleakage,m³/m²s qsi+se qse
qsi qsie
4 HYGROTHERMAL MEASUREMENTS 4.1 General
The measurements in the wall and roof sections started in October 1994 and lasted till June 1998. The logging system stores hourly mean values. What has been measured in each wall and roof element can be summarized as follows:
• Timber frame wall sections (12 sections): 8 moisture contents (bottom/top plate and studs) and 8 temperatures (bottom/top plate, studs and surfaces).
• Timber frame roof sections (6 sections): 2-6 moisture contents (beams and plywood) and 10 temperatures (beams, plywood and surfaces).
• Wall and roof sections of aerated concrete (3 sections): 3 relative humidities and 3 temperatures in holes in the aerated concrete.
• Timber frame wall sections-special investigation of forced convection (2 sections). 8 moisture contents (bottom/top plate and studs), 8 temperatures (bottom/top plate, studs and surfaces), 3 relative humidities/temperatures (insulation cavity).
Moisture content in wood
The moisture content of wood members was measured by traditional pin electrode resistance measurements. The measurement sequence was as follows:
1. The measurement output (voltage) was converted to moisture content using the converter Delmhorst MT(G) 40. The converter employs the following calibration curve:
2 , 40 5
, 15 16
, 5 62
, 3 5
,
0 4 3 2
0 =− ⋅U + ⋅U − ⋅U − ⋅U+
u
where u0 is moisture content (weight%) not compensated for temperature and wood species and U is the output voltage (V).
2. u0 is then compensated for temperature with the following formula:
( )
[ ] [ ( ) ]
8 , 2
2 0
1 0,881 1,0056
8 , 2 000051 , 0 8 , 2 026 , 0 567 , 0
⋅ +
+
⋅ +
+
⋅
−
=u + t t t
u
where u1 is moisture content corrected for temperature t (°C).
3. u1 is then compensated for wood species, i.e. in this case spruce:
4 1 3
1 2
1
1 0,01023 0,001233 0,000045 5986
, 1 5476 ,
1 u u u u
u=− + ⋅ − ⋅ − ⋅ + ⋅
where u is the corrected moisture content.
The diameter of the metal electrodes used was 2 mm and they were covered with 0.5 mm plastic except for 10 mm at the tip. The MC electrodes intruded from the insulation cavity side until a depth so that the unprotected part of the electrode was located 4-14 mm beneath the opposite wood surface (studs, sill plate and bottom plate). The location of the moisture electrodes in a top plate is shown in figure 15. The location of the moisture electrodes in the plywood in the roof sections is shown in figure 16. To avoid that condensed water at the wood surface could be transported into the wood along the moisture electrodes, silicone was used to tighten between the electrode and wood (see figure 15 and 16). The distance between the electrodes in a pair was 25 mm. In the wood MC range 7-25% the accuracy of the MC measurements is estimated to be within ±2 %, while values outside of this range have higher levels of uncertainties.
148 mm
4 mm 10 mm
COLD WARM
SIDE 48 mm SIDE
Spruce Measurement area
10 mm 3 mm Moisture electrodes 3 mm 10 mm Silicone
Figure 15 Detail of top plate in a timber frame wall section showing location of two pairs of moisture electrodes.
Plywood
Measurement area
22 mm
Silicone
Moisture electrode 25 mm
Figure 16 Detail of plywood in timber frame roof section showing location of a pair of moisture electrodes.
Identification system for measurement points
Each measurement point is given a unique alphanumeric 5 sign number which clearly identify the location of the measurement point, se Table 13 and 14. Example: E3U22 is a moisture content (U) measurement in the outer part of the top plate (22) in section E3.
15mm 3 mm
3 mm
